jhe first picentr 5 - accueil | epicentre · a catalyst to stimulate your thinking further on how...

�.J"he First picentr

. ...

:: ::::::·. ..

:::: :

epi0e ntre +ni.=

:: :rn:EPIDEMIOt.OGIE • EPIDEMIOt.OGY : i' ;;;:

5

994WB£33

These abstracts are not to be quoted for publication 1

Foreword

Welcome to you all!

It is great to see you at the very first Scientific Day Epicentre is organizing in Africa.

Epicentre has provided epidemiological expertise to Médecins sans frontières (MSF) from its Paris headquarters in France for 30 years and has had a base in

Uganda for more than 20 years and in Niger for almost a decade. Epicentre Uganda is located at the heart of Mbarara University of Science and Technology

(MUST) and Mbarara Regional Reference Hospital (MRRH). At this moment we take the opportunity to thank the University leadership and the ministry of

Health that allowed Epicentre to set up a permanent home on its campus from where great work through collaborative effort has been carried out.

Epicentre conducts research and training in support of MSF’s goal of providing medical care in areas where people are affected by conflicts, epidemics,

disasters, or are otherwise excluded from health care. In these contexts medical doctors do not always have tools to respond adequately to the health

needs of the people. Innovation to adapt treatment, diagnostic tools, preventive measures and strategies are needed. These innovations have to be built on

evidence without which no policy advance can take place. This is why Epicentre was created and what it continues to do to this day.

Today we have gathered here as people from different walks of life including politicians, diplomats, academicians, service providers, etc., but for one

common reason; to forge a way for better delivery of health services. As Epicentre, we are proud to share with you research results that we hope will act as

a catalyst to stimulate your thinking further on how to continuously improve health policies and hope to explore with you future research perspectives for

Africa.

In the first session we explore research projects that investigate how HIV care strategies can be adapted and improved. This will be followed by a series on

how tuberculosis treatment can be safely simplified whilst preserving efficacy. The third session will focus on complex emergencies where epidemiological

data need to be made available rapidly in order to mount an effective response. Our last session highlights patients whose needs are neglected despite

great suffering. We hope that this last session will stimulate a discussion on health research topics for Africa today.

At the end of the day we will gather for a drink reception where we can continue to discuss today’s topics informally.

I hope you have an excellent day!

Dr Juliet Mwanga, Director Epicentre Uganda


Programme Serena Hotel. 5th July, 2017

8:00 - 9:00 Registration

Morning programme

9:30 – 10:00 Opening Ceremony

Juliet Mwanga-Amumpaire, Director Epicentre Mbarara Research Centre

Professor Celestino Obua, Vice-chancellor Mbarara University of Science and Technology

Dr Jane Aceng, Minister of Health Uganda

10:00 – 11:00 HIV/AIDS: Research into adapted strategies to improve access to HIV care

Moderator: Dr. Joshua Musinguzi, Program Manager Aids Control Program, Ministry of Health, Uganda

o Understanding virological failure in HIV positive adolescents: is it true treatment failure or poor adherence?

Birgit Schramm

o Fishermen and Fishing communities in East Africa: Most-at-Risk population of acquiring HIV infection. Results from a population-based

survey

Juan Burgos Soto

o Contribution of HIV Population Impact Studies to care strategies : experience of 6 MSF African projects

David Mamano Ongoing project: Lopinavir-based ART for HIVInfected childreNGlobally (LIVING study)

Julian Adong

11:00 – 11:30 Coffee break


o Rifatox: Is it safe to double the dose of rifampicin to shorten tuberculosis treatment duration?

Daniel Atwine

o Rifavirenz: Can high-dose rifampicin be an option to shorten tuberculosis treatment for tuberculosis and HIV co-infected patients?

Maryline Bonnet

o Kids cohort: Pitfalls of childhood tuberculosis diagnosis in high-burden and limited-resource countries.

Elias KumbaKumba

o Ongoing project: TB Speed: Improvement of childhood tuberculosis diagnosis using new technologies for use in low-level health care

facilities.

Eric Wobudeya

12:30-14:00 Lunch

Afternoon programme

14:00 – 15:00 Emergencies and displaced people: research to guide adapted and timely responses

Moderator: Jean-Luc Anglade, Head of Mission Uganda, MSF France

o Nigeria: Borno crisis, mortality and malnutrition in areas of MSF interventionSophie Masson

o Pneumococcal carriage pre- and post-PCV vaccination campaign in Adjumani refugee camp, UgandaDan Nyehangane

o Dihydroartemisinin piperaquine as intermittent preventive treatment for children in refugee campMatthew Coldiron

o Ongoing project: Health assessment and surveillance in refugee camps in Northern UgandaDenis Ardiet

15: 00 – 15:30 Tea break

11:30 – 12:30 Tuberculosis: Optimizing treatment to improve patient outcomes

Moderator: Dr. Frank Mugabe; Program manager NTLP, Ministry of Health, Uganda


17: 00 – 17:30 Closure session

Emmanuel Baron, Director Epicentre, introduces:

Dr. Elioda Tumwesigye: Minister of Science, Technology and Innovation

Juliet Mwanga-Amumpaire closes meeting

17:30 Drinks

15:30 – 17:00 Research off the beaten track: neglected research topics for neglected patients. Perspectives for future research.

Moderator: Philip Guerin, Professor of Epidemiology and Global Health, Oxford University

o Oral cholera vaccines: transforming old products in new solutions for neglected populationsFrancisco Luquero

o PSYCa: Developing a tool to identify young children with mental health problems FabienneNackers

o Efficacy and safety of a new heat stable rotavirus vaccine (ROSE)Amadou Matar Seck

o Antivenoms for snake bites: Combatting market failure and neglect with researchMatthew Coldiron

o Infection control in the era of antibiotic resistanceCéline Langendorf

HIV/AIDS: Research into adapted strategies to improve access to HIV care

Session 1


Understanding virological failure in adolescents living with HIV: Evidence from a cross-sectional assessment and a qualitative study

Birgit Schramm, Epicentre, France

Background

The number of adolescents living with HIV is rising in sub-Saharan Africa, with

this patient group experiencing poor treatment and health outcomes. We

assessed virological failure and drug resistance to first-line antiretroviral therapy

(ART), and qualitatively explored issues influencing adherence amongst

adolescents living with HIV (ALHIV) in Chiradzulu, Malawi.

Methods

A mixed-methods study was conducted between May-November 2016.

Quantitative: A cross-sectional assessment included 10-19 year olds receiving

first-line ART for ≥6 months. Plasma viral load (VL) was assessed and drug

resistance-genotyping performed if VL ≥500 copies/ml. Participants with VL

>1 000 copies/ml received counselling and a second VL test.

Qualitative: We explored individual and social influences on ART adherence

through in-depth interviews with 16 adolescents, 16 caregivers, seven

community members, six health workers, and eight group discussions with

ALHIV and HIV-negative adolescents.

Results

Quantitative: 409 adolescents (median age 13 years, 58% female) participated

after a median of 6.7 years (IQR: 3, 15) on ART (85% AZT-3TC-NVP, 11% TDF-3TC-

EFV). Thirty-two-percent of participants had VL ≥1 000 copies/ml; only 16%

supressed to <1 000 copies/ml after counselling. Resistance testing revealed

80% of virological failures were on ≤1 effective drug. Major resistance included

3TC (86%), NVP (93%) and EFV (75%).

Qualitative: Most adolescents reported difficulties adhering to ART despite

understanding the importance for their health. Adherence was undermined by

individual factors such as poor mental health and inadequate HIV status

disclosure, family factors including precarious caretaking arrangements,

community factors including stigma; and health services factors such as poor

patient-provider relations and overly strict treatment-taking instructions

Conclusions

Treatment failure and drug resistance were high amongst ALHIV in Chiradzulu.

A robust once-per-day first-line regimen and frequent VL monitoring should be

considered to support adherence and minimize accumulation of resistance.

Family-centred approaches are needed alongside youth-friendly health services

with tailored counselling and peer-support clubs to help adolescents thrive on

ART.

Treatment failure is high among HIV-positive adolescents in Chiradzulu. Multiple

social challenges undermine treatment adherence requiring tailored models of

care involving families, peers and communities.


Fishermen and Fishing communities in East Africa: Most-at-Risk population of acquiring HIV infection; Results from a population-based survey

Juan Burgos-Soto, Epicentre, France

Background

In East-Africa, fishing communities are considered HIV most-at-risk population but knowledge about the epidemic in these contexts is scarce. Objective: To estimate the HIV-prevalence rate among adults and children in 12 fishing communities surrounding Lake George and Edward, Uganda; and to assess the HIV cascade of care in these settings.

Methods

We conducted a household-based survey. Trained-nurses visited 890 randomly-selected households interviewing adults of 15-69 years old. Blood samples of HIV-positives were collected for viral load measures. Children <15 years old were eligible for testing only if parents were HIV-positive. Logistic regression models, adjusted on socio-demographic behavioral variables were used to identify factors associated with HIV testing and being HIV-positive, and factors associated with HIV-status unawareness and viral suppression among HIV-positive adults.

Results

Overall, 1 738 adults and 148 children were included. The HIV-prevalence rate among adults was 17.5% (95%CI: 15.8-19.4) and among HIV-exposed children 6.1% (95%CI: 3.1-11.4). HIV-prevalence rate was higher among women (20.9%;95%CI: 18.4-23.5) than among men (13.5%; 95%CI: 11.3-16.1). According

to occupation and sex, farmers had the highest HIV-prevalence rate among women (27.6%) and fishermen among men (18.7%). After adjustment, only fishermen remained with a 4-fold higher risk of being HIV-positive (aOR: 3.9; 95%CI: 1.6-9.4), compared to men of other occupations. Among HIV-negatives, 81.0% declared having had a test in the preceding 12 months. Among HIV-positives, 86.0% declared HIV-status awareness, 78.0% were on ART and 56.0% virally suppressed. Men had a higher risk of being untested than women (aOR: 2.2; 95%CI: 1.4-3.7) and being virally detectable (aOR: 6.6; 95%CI: 1.9-22.0).

Conclusions

HIV-prevalence rate in Ugandan fishing communities is high, particularly among women and fishermen. Although HIV testing and ART initiation rates are high, viral suppression rate remains poor, especially among men. Nevertheless, fishermen do not seem to have a lower access to care than other men. More HIV preventive interventions are needed in these settings, particularly targeting women and fishermen. Strengthening ART-retention, particularly among men, should be a priority in these settings.

HIV-prevalence rate in Ugandan fishing communities is high, particularly among women and fishermen. More HIV preventive interventions are needed as well as strengthening ART-retention, particularly among men


HIV impact in population surveys in Eastern and Southern Africa

David Maman, Epicentre, South Africa

Introduction

To achieve HIV treatment goals, data are needed on the impact of programs. Epicentre has conducted 6 district based population surveys in 5 countries to measure coverage and impact in HIV programs supported by MSF. The results measured progress toward 90-90-90 goals and guided priorities.

Method

A total of six cross-sectional population surveys were implemented in Ndhiwa (Kenya, 2012), Chiradzulu and Nsanje (Malawi, 2013 and 2016, respectively), Eshowe (South Africa, 2013), Gutu (Zimbabwe, 2016), and Kasese (Uganda, 2016). Using multistage cluster sampling, we recruited individuals aged ≥15 years living in 2 400 selected households in Gutu and 2 443 households in Nsanje, and individuals aged 15-69 years living in 828 households in Kasese. Individuals who agreed to participate were interviewed and tested for HIV at home. All participants who tested positive had their viral load measured, regardless of their ART status. In some studies, we also evaluated CD4 counts, HIV incidence as well as transmitted and acquired resistance.

Results

Among 34 386 adults eligible from 15 473 included households, 30 245 (88.0%) were included and tested for HIV+, ranging from 1 738 in Kasese (inclusion 95.9%) to 7 269 in Chiradzulu (inclusion 87.8%).

The overall HIV prevalence ranged from 12.1% (95%CI 11.2-13.0) in Nsanje to 25.2% (95%CI 23.6-26.9) in Eshowe. In each site, the prevalence was higher among men compared to women (p<0.01).

Overall progress toward the 90-90-90 target was: 59/67/83 in Ndhiwa, 77/84/91 in Chiradzulu, 86/94/86 in Gutu, 77/91/89 in Nsanje, and 86/89/68 in Kasese. Each study highlighted that men and HIV-positive individuals younger than 30 were less likely to be diagnosed compared to respectively women or older individuals (p<0.01).

Conclusion

These six surveys, with the exception of the first one conducted in Ndhiwa in 2012, showed overall high coverage and highlighted remaining gaps in the cascade of care. In all sites, coverage outcomes were better among women than men, and among older than younger adults, mostly because they were less likely to be diagnosed. Interventions in these settings should emphasize detection of undiagnosed men and young adults.

We performed population-based surveys in 6 sites in 5 African countries to

evaluate the coverage and impact of HIV programs. In these settings, we showed

that high levels of coverage are achieved. However, detection of undiagnosed

men and young adults needs to be reinforced.


Ongoing project: Lopinavir-based ART for HIV-Infected childreNGlobally (LIVING study)

Julian Adong, Mbarara University of Science and Technology, Uganda

Introduction

The revised WHO treatment guidelines reemphasizes the need to treat all HIV-

infected children below 5 years of age and, and initiate therapy with a

Protease Inhibitor (PI)-based regimen for children less than 3 years of age

regardless of prior exposure to antiretroviral in prevention of mother to child

transmission (PMTCT) programme . Few antiretroviral drugs are approved for

infants and toddlers and existing combination antiretroviral therapies (cARTs)

which typically combine Non-Nucleoside reverse transcriptase inhibitors

(NNRTI) Nevirapine (NVP) and Nucleoside reverse transcriptase (NRTI)

lamivudine (3TC) plus either Abacavir (ABC) or Zidovudine (AZT). They are not

optimal for newly infected infants due to their high viral loads (10 to 100 times

higher than older children); in addition they are often infected with viruses

already resistant to NVP, commonly used for PMTCT. For infants the only

alternative to NVP is the PI, Lopinavir, boosted with ritonavir; however the

available LPV/r liquid formulations for children taste very bitter, are difficult to

administer, and are unstable in tropical climates. This study is evaluating the

effectiveness, safety and acceptability of LPV/r pellets in addition to AZT/3TC

(or ABC/3TC) paediatric fixed dose combination tablet under routine treatment

conditions in HIV infected infants and young children who cannot swallow

tablets.

Methods

It is single arm, open-label, prospective, multicenter, multi-country phase IIIb

study among children with a past or current documentation of a confirmed

diagnosis of HIV-1 infection, eligible for treatment with LPV-based treatment,

weighing ≥3 and <25 kg at the time of enrolment and unable to swallow

tablets. The primary endpoint is treatment effectiveness at 48 weeks based on

a composite endpoint of virologic response <1 000 copies/ml, being alive and

on study drug. Patients are followed-up for 24 months.

Status of study

At the Epicentre site recruitment of participants started on 10th May 2016

and so far 68 patients have been enrolled to date.

Existing combination antiretroviral therapies for children are complex, and not

optimal for newly infected infants with very high viral loads. There is need to

study Lopinavir boosted with ritonavir in pellet form as the available paediatric

liquid formulations have a very bitter taste and are unstable in tropical climates.

Tuberculosis: Optimizing treatment to improve

patient outcomes

Session 2


Is it safe to double the dose of rifampicin to shorten tuberculosis treatment duration?

Daniel Atwine Epicentre Uganda

Background

The current 6-month TB treatment regimen is still too long. Use of higher doses

of rifampicin(R) might ensure faster lung sterilization with subsequent reduction

in treatment duration. However, R can induce liver toxicity and safety data using

high dose rifampicin are lacking. We assessed whether increasing the dose of

rifampicin (R) from 10 mg/kg to 15 or 20 mg/kg, results in an increase in grade

3 or 4 hepatic adverse events and/or serious adverse events (SAE).

Methods

In a Phase IIB, open-label randomized controlled trial in Bolivia, Peru, and

Uganda, 300 HIV negative patients with newly diagnosed, smear positive, drug

susceptible pulmonary tuberculosis were randomly assigned to one of three

regimens containing ethambutol, isoniazid, rifampicin, and pyrazinamide daily

for 8 weeks followed by isoniazid and rifampicin daily for 18 weeks. The

regimens differed only by the R dose during the first 16 weeks: 1 control regimen

with R at 10mg/Kg and 2 study regimens with R at 15mg/Kg (R15) and 20mg/Kg

(R20). Serum alanine transferase (ALT) measurements were carried out at

regular intervals. Eight week culture conversion was used as surrogate marker

for treatment efficacy.

Results

There were 7 grade 3 increases in ALT levels, 1/100 (1.0%) in R10, 2/100 (2.0%)

in R15 and 4/100 (4.0%) in R20 regimens respectively (trend test p=0.15). R was

discontinued due to liver toxicity in 1 patient (R15). There were no grade 4 ALT

increases. There was a non-significant increase in culture conversion rate with

increasing rifampicin dosage, 75% (69/92) R10, 82.5% (66/80) R15, and 83.1%

(76/91)) R20, (p=0.16).

Conclusions

There was no significant increase in liver toxicity when rifampicin dose increased

from 10mg/kg to 15mg/kg or 20mg/kg.

The use of high-dose rifampicin in shortening of TB treatment duration from 6

months to 3 or 4 months can be possible if its safety is assured. Results of

RIFATOX trial, show no significant increase in drug-induced Liver toxicity even

with doubling of the normal dose of rifampicin.


Can high-dose rifampicin be an option to shorten tuberculosis treatment for tuberculosis and HIV co-infected patients?

Maryline Bonnet, Epicentre, on behalf of IRD UMI 233 TransVIHMI - UM – INSERM U1175, France

Background

Ongoing trials are evaluating high-dose rifampicin (R) regimens to shorten

tuberculosis (TB) treatment duration. The risk of drug interaction with some

antiretroviral precludes the inclusion of HIV-infected patients. We assessed the

effect of high-dose R on the efavirenz (EFV) metabolism in co-infected patients.

Methods

RIFAVIRENZ was a phase-2, randomized, open-label trial conducted in Uganda

between 2014 and 2017. Pulmonary TB and antiretroviral therapy (ART)-naïve

patients were randomized to 2-study regimens (SR) using high-dose R (20mg/Kg)

with ART initiation 2-4 weeks later with 600mg/day (SR1) or 800mg/day (SR2)

EFV; or to 1-control regimen (CR) using R10mg/Kg and EFV600mg/day. At 8

weeks, all patients were switched to standard R and EFV doses. All patients had

intensive pharmacokinetic sampling 4 weeks after EFV-R co-administration, and

4 weeks after R discontinuation. HIV and TB treatment response and safety were

monitored.

Preliminary results

Of 97 included patients (SR1: 31; SR2: 33; CR: 33), 26.8% were females and

median age, weight and CD4 count were 33 years, 53.6 kg and 141 cells/L,

respectively. Under R, the median of the EFV minimum concentration (C24) was

1188, 1064 and 1078ng/mL for SR1 (N=27), SR2 (N=30) and CR (N=28),

respectively. Five (18.5%), 6 (20.0%) and 8 (28.6%) patients had C24 < 750ng/mL.

At 12 weeks post-ART initiation, 92.6%, 86.2% and 92.6% of patients had HIV

viral load < 400 copies/mL. Week 8 TB culture conversion was 88.5% (SR1), 88.9%

(SR2) and 90.3% (CR). During first 8 weeks, 6 (2 per arm) and 4 patients (SR1: 1;

SR2: 2; CR: 1) had transaminase increase > grade 3 and neuropsychiatric events

> grade 2, respectively.

Conclusions

Doubling the R dose does not seem to affect the EFV concentrations. These

preliminary results need confirmation with the comparison of the EFV

pharmacokinetics parameters with and without R.

Trials evaluating high-dose rifampicin short regimens exclude HIV-infected

patients due to risk of drug interaction with antiretroviral. Based on the

preliminary RIFAVIRENZ trial results, doubling the rifampicin dose does not seem

to affect the efavirenz concentrations.


Kids Cohort: Pitfalls of childhood tuberculosis diagnosis in high burden and limited resource settings.

Elias Kumbakumba, Epicentre. Uganda

Background

Diagnosing tuberculosis in children is very challenging, especially in high-burden

resource-limited settings. We report the diagnostic findings of a cohort of

children with clinical suspicion of tuberculosis at the Mbarara Regional Referral

Hospital.

Methods

All children suspected of tuberculosis aged between one month and 14 years

received a comprehensive and standardized clinical and biological evaluation

including XpertMTB/RIF assay and tuberculosis culture and use of induced

sputum in children unable to expectorate. After initial assessment, children with

any positive tuberculosis bacteriological test, chest x-ray or clinical presentation

that were suggestive of tuberculosis disease were started on treatment. Stools

collected from children on treatment were tested with XpertMTB/RIF. At the

end of the study, two independent experts reviewed patients’ files to classify

cases using the 2012 standard case definitions of intra-thoracic tuberculosis.

Results

Of 392 children enrolled, 45.4% were female, 58.1% were younger than 5 years

old, 30.1% were HIV-infected and 19.4% had a tuberculosis contact history

within the last year. Children < 2 years presented more frequently with severe

malnutrition (26.4% vs 5.3%) and were more likely to be diagnosed with a

tuberculosis suggestive chest X-ray (50.8% vs 29.2%) compared to children from

other age groups. A quarter (25.8%) of the children were tuberculin skin test

positive.

Nineteen children (4.8%) were confirmed tuberculosis cases: 18 were either

culture or Xpert positive for MTB on respiratory samples and 1 was positive on

extra-pulmonary sample. Xpert MTB/RIF on stool had a sensitivity of 57.1% and

specificity of 98.2%. Using the standard tuberculosis case definitions, a total of

58/373 (15.4%) children were classified as confirmed or probable tuberculosis.

In total, 144 (36.7%) were started on treatment.

Conclusion

This study confirms the difficulties with diagnosing childhood tuberculosis. The

majority of cases were diagnosed without confirmation of tuberculosis despite

the use of exhaustive diagnostic tests and optimised specimen collection. Rapid,

non-sputum based and highly sensitive tests are still needed for diagnosis of

childhood tuberculosis. Meanwhile, every opportunity should be taken to

improve tuberculosis diagnosis especially among children presenting with

severe clinical conditions.


TB Speed: Improvement of childhood tuberculosis diagnosis using new technologies for use in low-level health care facilities.

Eric Wobudeya, MUJHU care ltd & Mulago National Referral hospital, Uganda

Background

Notification of tuberculosis in children is still low. This likely to be due to the

paucibacillary nature of childhood tuberculosis, the difficulty to obtain sputum,

the low diagnostic yield of existing tests, weak chest radiograph services, lack of

systematic screening at all health facility entry points, and absence of point-of-

care tests. Improved case detection and access to treatment for children with

tuberculosis is a key step to reach the goal of zero death from tuberculosis.

Objective

The TB-SPEED project goal is to contribute to the reduction in childhood

mortality from tuberculosis by delivering an available, feasible, cost-effective,

and decentralized childhood tuberculosis diagnostic approach to enhance case-

finding and access to treatment.

Project implementation plan

This is a four-year UNITAID-funded project implemented in seven countries

(Cambodia, Cameroon, Côte d’Ivoire, Mozambique, Sierra Leone, Uganda, and

Zambia) to start in September 2017 with three major patient-centred outputs

and three technical support outputs building evidence for future scale-up.

The project will innovatively validate a decentralized diagnostic approach using

battery-operated GeneXpert OMNI and Xpert ultra cartridge to increase access

to tuberculosis diagnostics. It also proposes a simplified nasopharyngeal

aspirate sample collection method through the use of battery-operated suction

machines that are suitable for rural areas. The project also aims to optimize stool

processing without centrifugation allowing stool Xpert testing at lower-level

health care facilities. The approach will also include optimized screening, clinical

and radiological diagnosis through training, mentorship, the use of digital

radiography, and establishment of quality control of X-ray reading.

Project outputs

1. New decentralized childhood tuberculosis diagnostic approaches tested at

district health system level

2. Evaluation of an early tuberculosis detection strategy in children with

severe pneumonia

3. Validation of diagnostic tools and algorithms in highly vulnerable groups

with presumptive tuberculosis, specifically HIV-infected and severely

malnourished children

4. Identification of optimized, suitable, and affordable specimen collection

and processing methods for childhood tuberculosis diagnosis in

resource-limited countries

5. Evaluation of cost-effectiveness of the proposed diagnostic approaches

6. Dissemination, communication and stakeholders’ engagement

The TB-SPEED project aims to increase case detection and reduce

mortality of children with tuberculosis through decentralization of diagnosis systematic and screening for tuberculosis of children with HIV infection severe

malnutrition, or severe pneumonia.

Emergencies and displaced people:

research to guide adapted and timely responses

Session 3


Borno crisis, Nigeria. Mortality and nutrition in areas of MSF intervention

Sophie Masson, Epicentre, France

Background

Since 2013, the Boko Haram insurgency and military operations have led to mass

population displacement in northeastern Nigeria. From June 2016, MSF was able

to operate inside and to lesser extend outside of Maiduguri. We present a series

of surveys and surveillance results between July 2016 and May 2017, to describe

the health and nutrition status of the population and follow the situation over

time.

Methods

In Banki camp, 4 retrospective mortality surveys, using systematic sampling

were undertaken between July and December 2016, coupled with malnutrition

assessments. In Maiduguri, prospective surveillance of mortality, population

size, and malnutrition was carried out in 12 camps in 2016. Retrospective

mortality surveys and nutritional assessments were carried out in two unofficial

camps between September and October 2016, using exhaustive and systematic

sampling. A survey using spatial sampling covering the urban area of

Maiduguriexcluding campswas carried out in November 2016. Cross-

sectional population based surveys using spatial cluster sampling were carried

out in the catchment areas of MSF nutrition programme to estimate prevalence

of malnutrition and programme coverage in May 2017.

Results

In Banki, the initial retrospective mortality and rapid malnutrition screening in

July 2016 demonstrated an extremely critical situation. The subsequent surveys

showed a rapid decrease of both mortality and malnutrition. The surveys in the

two unofficial camps in Maiduguri were also indicative of a critical situation,

while the survey of the overall urban areas showed low mortality and

malnutrition for both host community and internal displaced people in the

community. The surveys in the Maiduguri MSF ambulatory treatment feeding

centre’s (ATFC) catchment areas showed low prevalence of malnutrition.

Nevertheless, admissions to ATFCs remained very high.

Conclusion

The situation in camps assessed outside of Maiduguri were critical, but showed

a rapid improvement, despite challenges of accessibility. In Maiduguri, where

accessibility to and from the population is less challenging, Overall indicators

were much better but there remained pockets of vulnerability”. The high

number of patients presenting to services is better explained by a large number

of people in the area rather than by high prevalence or good coverage.

Following mass displacement of population in Borno state, Nigeria, and

subsequent MSF emergency response, various retrospective mortality surveys

and rapid nutritional assessments were carried out in order to describe and

follow over time the extent of the crisis and its evolution in the areas of MSF

intervention. Evaluation of dense urban areas remains a challenge.


Pneumococcal carriage and serotypes distribution pre- and post- PCV vaccination campaign in Adjumani refugee camps, Uganda

Dan Nyehangane, Epicentre, Uganda

Background

In July-September 2014, MSF conducted a mass vaccination campaign (MVC)

among children aged between 6 weeks and 23 months in Adjunami settlements

using pneumococcal conjugate (PCV-10) vaccines. We aimed to examine the

impact of the MVC on pneumococcal carriage and serotypes circulation.

Methods

Three nasopharyngeal (NP) pneumococcal carriage household surveys

(respectively in July 2014, March and June 2015) were carried out among

residents (all ages). Streptococcus pneumoniae was cultured from NP swab

specimens. All pneumococcal isolated were serotyped using the Quellung

method in Epicentre Mbarara-Uganda Laboratory, in order to describe

pneumococcal serotyping distribution in Adjumani population before and after

the PCV mass vaccination campaign. In addition, MSF conducted a vaccination

coverage cluster survey among children aged between 6 weeks and 23 months

in October 2014.

Results

Among children in the age range for vaccination, 96% (95%CI 94-98%) received

at least one dose of PCV-10 and 43% (95%CI 39-47%) received 3 doses of PCV-

10. Overall pneumococcal carriage increased from 58% (95%CI 56-61%) in the

first survey to 67% (95%CI 64-69%) in the third, and from 86% (95%CI 83-90%)

to 92% (95%CI 90-94%) among children younger than 24 months. Vaccines

serotypes decreased from 37% (95%CI 34 – 40%) to 15% (95%CI 14-18%) overall,

and from 49% (95%CI 44-54%) to 17% (95%CI 14-21%) among children less than

24 months.

Conclusion

Following the PCV-10 mass vaccination campaign, NP carriage increased. A

significant and rapid replacement of vaccines serotypes by non-vaccines

serotypes was observed in all age groups, including non-vaccinated age-groups.

Non-vaccines serotypes are expected to be less invasive. Improvement of

surveillance of invasive pneumococcal disease is required, together with

serotypes circulating in order to monitor the impact of PCV mass vaccination.

Nasopharyngeal pneumococcal carriage increased among the overall population

of Adjumani settlements, following a PCV-10 mass vaccination campaign

organized between July and September 2014. A significant and rapid

replacement of vaccines serotypes by non-vaccine serotypes was observed in all

age groups, including non-vaccinated age-groups.


Dihydroartemisinin-piperaquine as intermittent preventive treatment for children in refugee camp

Matthew Coldiron, Epicentre, France

Background

Northern Uganda hosts a large population of refugees from South Sudan, and

malaria is one of the major health problems in the area. In 2015, intermittent

preventive treatment for malaria (IPTc) was implemented in two refugee

camps among children aged six months to 14 years.

Methods

Three distributions of dihydroartemisinin-piperaquine (DP) were conducted at

eight-week intervals. The first dose was directly administered at IPTc

distribution sites and the second and third doses were given to caregivers to

administer at home. A multi-faceted evaluation was implemented, including

coverage surveys, malaria prevalence surveys, reinforced surveillance, and

pharmacovigilance.

Results

Programme coverage exceeded 90% during all three distributions with a total

of 40 611 participants. Compared to same period during the previous year

(only available data), the incidence of malaria in the target populations was

reduced (IRR 0.73, 95% CI 0.69-0.77 among children under five years old; IRR

0.70, 95% CI 0.67-0.72 among children aged five-14 years). Among those not

targeted for intervention, the incidence between the two years increased (IRR

1.49, 95% CI 1.42-1.56). During the intervention, estimates of the prevalence

of parasitemia (by microscopy or PCR) ranged between 12.9-16.4%, with the

highest prevalence among children aged five-14 years. This increased to 25.3%

(95% CI 22.1-28.8) eight weeks after the final distribution. A total of 57 adverse

events were reported during the intervention period, including 1 severe

adverse event (death from varicella). Adverse events were of mild to moderate

severity, and were mainly dermatologic and gastrointestinal.

Conclusion

This is the first documentation of an IPTc programme in a refugee camp. The

positive impact of DP on the incidence of malaria, together with its favourable

safety profile, should lead to further use of IPTc in similar settings. Expanding

coverage groups and decreasing intervals between distributions might provide

more benefit, but would need to be balanced with the operational implications

of a broader, more frequent distribution schedule.

Preventing malaria in vulnerable populations is a major challenge. The use of

DHA-Piperaquine was an effective and safe strategy for intermittent preventive

treatment in Uganda


Health assessment and surveillance in refugee camps in Northern Uganda

Denis Ardiet, Epicentre, Paris

Background

Following escalation of violence in South Sudan in July 2016, thousands of

refugees crossed the border with Uganda. The Ugandan government settled

them into the Bidibidi, and subsequently also the Imvepi settlement. Because

they were highly dependent on humanitarian aid, a baseline health and

mortality survey assessed their health status. This was followed by the

implementation of two different health surveillance systems allowing weekly

reporting of most basic health indicators.

Methods

Households were randomly selected by spatial sampling, and household

structures were assessed. Nutrition status for children <5 years was evaluated

using Mid-Upper Arm Circumference (MUAC) and edema assessment.

Retrospective mortality used a 5-months recall period. Two different weekly

surveillance systems were implemented, one following the Ugandan community

health system and another “lighter” system, focusing on mortality and most

epidemic diseases. A second survey was performed for new arrivals to collect

more information about mortality in South Sudan and during their journey.

Results

A total of 1 018 heads of household accepted to participate in the baseline

survey. The population was found to be very young, and split households were

frequent with 20% of that population missing; 32% of households were headed

by women. Many households also lacked crucial non-food items. In the

settlements, malnutrition and mortality appeared to be below emergency

thresholds but delays in food distribution were frequent. On the other hand,

crude mortality rates were found to be high in South Sudan with many violent

deaths recorded. Surveillance systems revealed small bursts of bloody diarrhea

and malnutrition pockets in both settlements.

Conclusions

Our assessments reflected high levels of violence in South Sudan. In the Uganda

settlements, health indicators were under control but still need to be

monitored.

Health assessment and surveillance among South Sudanese refugees are

essential activities, as access to food, water and health services remains fragile.

Research off the beaten track:

neglected research topics for neglected patients.

Perspectives for future research

Session 4

These abstracts are not to be quoted for publication

17

Oral cholera vaccines: transforming old products into new solutions for neglected populations

Francisco Luquero, Epicentre, Switzerland

Cholera vaccines have only recently become a control tool despite the fact that

early prototypes of the currently prequalified cholera vaccines exist since the

1880s. Key research conducted by Médecins Sans Frontières and Epicentre has

shown that mass campaigns using oral cholera vaccines (OCV) are feasible in

different settings (humanitarian crises, outbreaks and endemic countries), well

accepted by the population, and that the vaccines are a safe and effective tool

for prevention and response. In addition, we recently demonstrated that short-

term vaccine protection can be achieved with a single dose of vaccine, which is

a major logistical advantage in response to outbreaks. These key data, along

with the creation of a global stockpile managed by the WHO and dedicated

funding have led to the doubling of the number of OCV doses delivered

worldwide each year since 2012.

However, the significant public health benefit from OCV has not yet been

realized because of cost, availability (supply) and logistical constraints. These

include a recommendation for two doses, with the second delivered 14 days

after the first, high packing volume and cold chain requirements. These factors

make vaccine delivery costly and challenging in certain settings. In addition, the

complex vaccine formulation translates in suboptimal production, which limits

vaccine availability and impedes price reduction (~1.85USD per dose), making

the total cost of one fully vaccinated person approximately 6USD.

An improved new generation cholera vaccine could lift these barriers and will

bring substantial public health benefits. Any new vaccine should be easier to

produce, cheaper, heat-stable and have a reduced storage volume. As one of

the main responders to cholera epidemics, MSF could play an important role in

pushing for improved vaccines. With better cholera vaccines, used in

conjunction with WaSH measures, many more lives could be saved and perhaps

the elimination of cholera outbreaks could be foreseen in a near future.

The potential impact of cholera vaccine has not fully materialized because of

limitations of current products. New vaccines should be easier to produce,

cheaper, heat-stable and with reduced storage volume. MSF and Epicentre

could play an important role in pushing for such improved vaccines.


PSYCa: Developing a tool to identify young children with mental health problems

Fabienne Nackers, Epicentre, Belgium

Background

In low-resource settings, the lack of mental health human resources and the

absence of cross-culturally validated screening instruments jeopardize the

implementation of mental health care, especially for very young children. We

aimed to develop and validate a cross-cultural general tool, the PSYCa 6-36, to

screen for psychological difficulties in children aged 6 to 36 months.

Methods

A primary validation of the PSYCa 6-36 was conducted in Kenya (n=319 children

aged 6 to 36 months; 2014), followed by three secondary validations (n=215,

Kenya, 2014; n=189, Cambodia, 2015; n=182, Uganda, 2016). After standardized

translation procedures, lay interviewers administrated the PSYCa 6-36 in local

languages to the children’s caregivers at home. We assessed the psychometric

properties of the tool and its external validity against a gold standard (i.e. clinical

global impression severity [CGIS] score rated by a psychologist after clinical

assessment).

Results

The internal consistency of the PSYCa 6-36 was acceptable (Cronbach’s alpha

ranged from 0.61 to 0.74) and its temporal reliability was very good (intra-class

correlation coefficient [ICC] ≥0.80). The inter-rater reliability was acceptable to

good (ICC ranged from 0.60 to 0.83) as well as the external validity (area under

the curve ranged from 0.63 to 0.80). The prevalence of CGIS scores ≥1, indicating

mental health difficulties according to the psychologist, was 5.1% in Kenya, 8.7%

in Cambodia and 10.5% in Uganda.

Conclusion

The results of this study show that the PSYCa 6-36 is a promising screening tool

for young children. Once adapted to the local context, the PSYCa 6-36 was easy

and quick to use for trained non-specialists. The PSYCa 6-36 also increased the

awareness on children’s psychological difficulties and the importance of early

recognition to prevent long-term consequences. Further use and validation of

the tool in settings with higher prevalence of psychological difficulties will help

to refine the scale.

The PSYCa 6-36 is a useful tool to screen for psychological difficulties among

children aged 6 to 36 months.


Efficacy and safety of a new heat stable rotavirus vaccine (ROSE)

Amadou Matar Seck , Epicentre, Niger

Background

Every year, rotavirus gastroenteritis causes about one-third of deaths due to

diarrhea among children under 5 worldwide, most of which occur in sub-

Saharan Africa. Infection can be prevented with a vaccine, but a global

shortage of the vaccine and the need for cold-chain transport presents delivery

challenges.

Methods

We conducted a double-blind, randomized, controlled trial in Niger to evaluate

the efficacy and safety of BRV-PV (Serum Institute of India, Pvt Limited), a live,

oral, heat-stable rotavirus vaccine. Healthy infants received three doses of

vaccine or placebo at 6, 10, and 14 weeks of age. Episodes of gastroenteritis

were assessed through active and passive surveillance. The primary endpoint

was vaccine efficacy from 28 days post-dose 3. Assuming a 2% attack rate, a

50% true vaccine efficacy, and 20% participant non-assessibility, 117 cases (78

unvaccinated and 39 vaccinated) were required to establish 50% true vaccine

efficacy. All serious adverse events, including intussusception, and adverse

events were assessed using facility- and home-based surveillance.

Results

Among the 3 508 infants included in the per-protocol efficacy analysis (1 780 in

the vaccine group and 1 728 in the placebo group), there were 31 and 87 cases

of severe rotavirus gastroenteritis in the vaccine and placebo groups,

respectively (2.14 and 6.44 cases per 100 person-years), resulting in a vaccine

efficacy of 66.7% (95%CI 49.9–77.9). Similar efficacy was seen in the intention-

to-treat analyses. There was no difference in the risk of adverse events (68.7%

vaccine and 67.2% placebo) or serious adverse events (8.3% vaccine and 9.1%

placebo), including death (n=27 vaccine and n=22 placebo). No child had

confirmed intussusception.

Conclusion

We found BRV-PV to protect against severe rotavirus gastroenteritis among

infants. BRV-PV does not require refrigeration and represents an important

option to prevent morbidity and mortality among the most vulnerable

Rotavirus gastroenteritis leads to about one-third of diarrheal disease deaths in

children. We found BRV-PV to protect against severe rotavirus gastroenteritis

among infants in Niger


Antivenoms for snake bites: Combatting market failure and neglect with research

Matthew Coldiron, Epicentre, France

Venomous snakes are widely distributed, but most snakebites occur in rural

areas of the tropics. Although data are notoriously poor, in Africa, there are an

estimated 1 million snakebites, 500 000 envenomings, and up to 30 000 deaths

annually.

The venom of any single snake may contain more than 100 different toxins and

enzymes, so the clinical result for victims can differ as well. Three major

families of toxicities exist: local cytotoxicity (tissue necrosis), neurotoxicity

(paralysis), and hematotoxicity (pro-coagulant and anti-coagulant).

Snakebite patients with evidence of envenoming should be promptly treated

with antivenom. The choice of antivenom depends on the local epidemiology

of snakebites and the availability of antivenoms. If given with minimal delay,

antivenoms can drastically reduce mortality associated with snakebite,

particularly for hematotoxic envenomings, which typically cause death many

hours or days after the snakebite.

Antivenom production is onerous and has changed little since the 19th century.

In brief, venom is harvested from snakes and small amounts are injected into a

mammal (usually horses). After several months, the equine serum is collected

and purified, and anti-venom specific antibodies are collected, and sometimes

digested into fragments. Some commercially available antivenoms are species-

specific, others are polyvalent.

FAV-Africa (Sanofi Pasteur) is a polyspecific antivenom which covers 10

different species of snakes. It has been the primary antivenom used by MSF

across Africa for many years. Its production ended in 2013, and the last doses

produced expired in June 2016. Several new antivenoms have entered the

market in recent years, but with insufficient data supporting their safety and

efficacy. In several settings, MSF is facing challenges over the choice of

antivenom (or antivenoms) to use to replace FAV-Africa. Epicentre is

conducting studies in the Central African Republic and South Sudan to evaluate

and document the use of new products.

Snakebite is a major health problem in rural Africa. Although there are several

antivenoms available, there is insufficient data to support their safety and

efficacy.


Infection control in the era of antibiotic resistance

Céline Langendorf, Epicentre, France

Prescribing antibiotics without considering the risk of resistance is no longer

possible today. First, antibiotic resistance is rising to high levels across the world,

threatening ability to treat infectious diseases, even the most common ones.

Second, antibiotic resistance is accelerated by the misuse of antibiotics and poor

infection prevention and control.

In hospitals and community settings, ongoing research takes a holistic approach

by addressing different aspects: diagnosis, treatment and prevention. We are

conducting several descriptive studies with the objective of filling in data gaps

on the etiologies of bacterial infections, antibiotic resistance and to adapt

therapeutic protocols.

The presentation will be based on, first, several etiology studies in children in

sub-Saharan Africa; and next on documentation of how antibiotic are

prescribed. This provides an insight into gaps and opportunities to improve

antibiotic formulation, especially in pediatrics. The use of antibiotic prophylaxis

can increase the transmission of multi-drug resistant organisms within the

community. Finally, research on infection prevention and control and antibiotic

resistance is essential to improve the treatment protocols and reduce the

impact and spread of resistance

Antibiotic resistance is rising to high levels across the world, threatening ability

to treat infectious diseases. We discuss how research on infection prevention

and control and antibiotic resistance can help improve the treatment protocols

while reducing the spread of resistance.


Notes

Epicentre 8 rue Saint Sabin

75011 Paris France

Tel : +33 1 40 21 55 55

Epicentre Mbarara Research CentreP.O Box 1956

MbararaUganda

Email: [email protected]

www.epicentre.msf.org

jhe first picentr 5 - accueil | epicentre · a catalyst to stimulate your thinking further on how...

Documents